Papers by Matthew Lease
Promoting Constructive Deliberation: Reframing for Receptiveness (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Current methods for promoting pro-social discussion and debate online are limited. |
| Approach: | They propose automatic reframing of disagreeing responses to signal receptiveness to a preceding comment. |
| Outcome: | The proposed framework can be used to promote constructive debate and debate online. |
Benchmark Transparency: Measuring the Impact of Data on Evaluation (2024.naacl-long)
Copied to clipboard
| Challenge: | In this paper, we quantify the impact that data distribution has on the performance and evaluation of NLP models. |
| Approach: | They propose to use disproportional stratified sampling to measure the data distribution across 6 different dimensions to quantify model performance. |
| Outcome: | The proposed framework measures the data distribution across 6 different dimensions and shows that it is statistically significant and predicts model performance. |
Improving the Distributional Alignment of LLMs using Supervision (2026.acl-long)
Copied to clipboard
Gauri Kambhatla, Sanjana Gautam, Angela Zhang, Alexander Liu, Ravi Srinivasan, Junyi Jessy Li, Matthew Lease
| Challenge: | Existing work to evaluate LLMs' alignment with human values and opinions has a key shortcoming. |
| Approach: | They propose to add supervision to LLMs to improve alignment with diverse populations . they find that supervision improves alignment across public health, public opinion, values and beliefs . |
| Outcome: | The proposed method improves the alignment of LLMs with diverse populations on subjective questions. |
ProtoTEx: Explaining Model Decisions with Prototype Tensors (2022.acl-long)
Copied to clipboard
| Challenge: | Neural models for NLP have yielded significant gains in predictive accuracy across tasks. |
| Approach: | They propose a white-box NLP classification architecture based on prototype networks . they propose an interleaved training algorithm that faithfully explains model decisions . |
| Outcome: | The proposed model matches BART-large and exceeds BERTlarge on propaganda detection tasks. |